Reinforcement Learning, not Supervised Learning, Can Lead to Insight.
Arata NonamiHaruaki FukudaYoshiyuki SatoKazuyuki SamejimaKazuhiro UedaPublished in: CogSci (2018)
Keyphrases
- reinforcement learning
- supervised learning
- function approximation
- unsupervised learning
- kernel based learning
- learning problems
- learning algorithm
- state space
- reinforcement learning algorithms
- training data
- data sets
- temporal difference
- machine learning
- model free
- training examples
- learning tasks
- training set
- information systems
- action selection
- supervised classification
- optimal control
- neural network
- transfer learning
- real time
- learning process
- dynamic programming
- active learning
- optimal policy
- statistical learning
- labeled data
- dimensionality reduction
- multi agent
- robot control